Medical physics · 2025
Sensitivity analysis of key CT simulation parameters and their impact on quality of virtual images.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Montero Isabel, Samei Ehsan, Abadi Ehsan
Аннотация
BACKGROUND: Trials of medical imaging systems using clinical cases are often impractical motivating the use of virtual imaging trials (VITs). To emulate reality, VITs need models of imaging systems, whose realism depends on simulation parameter choice.
PURPOSE: To evaluate how simulation parameter choice impacts image quality of virtual Computed Tomography (CT) images.
METHODS: Utilizing a validated CT simulator (DukeSim, Duke University), a size-variant and an anthropomorphic computational phantom were imaged emulating a clinical scanner (SOMATOM Force, Siemens Healthineers) and adult abdomen protocol. Simulations varied parameters associated with x-ray source, phantom, and detector: source and detector subsampling (1-5, each side), anode heel effect severity (0%-40%), x-ray spectrum shape (± 5 keV effective energy shift), phantom voxel size (0.1-0.5 mm), and detector crosstalk magnitude (0%-50%, per dimension). Simulations were reconstructed employing vendor-specific software (ReconCT, Siemens Healthineers). Sensitivity to parameter choice was assessed using modulation transfer function (MTF, spatial resolution), noise magnitude, noise power spectrum (NPS, noise texture), and mean CT number.
RESULTS: MTF was sensitive to subsampling and crosstalk, with average changes in half-value frequency (f50) up to 0.055and 0.043 mm-1, respectively. NPS was sensitive to crosstalk, with average peak frequency (fpeak) changes of 0.045 mm-1. Mean CT number of the bone insert varied ± 103.8 HU with spectrum shifts. In anthropomorphic simulations, trabecular bone CT number varied by 26 ± 13 HU.
CONCLUSION: Simulation parameter choice impacts virtual CT image quality. Certain parameters exert greater influence and may need extra attention when developing and utilizing VITs.
PURPOSE: To evaluate how simulation parameter choice impacts image quality of virtual Computed Tomography (CT) images.
METHODS: Utilizing a validated CT simulator (DukeSim, Duke University), a size-variant and an anthropomorphic computational phantom were imaged emulating a clinical scanner (SOMATOM Force, Siemens Healthineers) and adult abdomen protocol. Simulations varied parameters associated with x-ray source, phantom, and detector: source and detector subsampling (1-5, each side), anode heel effect severity (0%-40%), x-ray spectrum shape (± 5 keV effective energy shift), phantom voxel size (0.1-0.5 mm), and detector crosstalk magnitude (0%-50%, per dimension). Simulations were reconstructed employing vendor-specific software (ReconCT, Siemens Healthineers). Sensitivity to parameter choice was assessed using modulation transfer function (MTF, spatial resolution), noise magnitude, noise power spectrum (NPS, noise texture), and mean CT number.
RESULTS: MTF was sensitive to subsampling and crosstalk, with average changes in half-value frequency (f50) up to 0.055and 0.043 mm-1, respectively. NPS was sensitive to crosstalk, with average peak frequency (fpeak) changes of 0.045 mm-1. Mean CT number of the bone insert varied ± 103.8 HU with spectrum shifts. In anthropomorphic simulations, trabecular bone CT number varied by 26 ± 13 HU.
CONCLUSION: Simulation parameter choice impacts virtual CT image quality. Certain parameters exert greater influence and may need extra attention when developing and utilizing VITs.